Exploring Audience Segmentation in Food Safety Social Media

Background: Social media is a key dissemination channel for public health information. Public health professionals must consider how various underlying factors shape the effectiveness of disseminating health messages on social media to reach key audiences.

Program background: CDC’s Division of Foodborne, Waterborne, and Environmental Diseases (DFWED) uses various communication activities, including social media, to increase the impact of foodborne illness prevention messages and improve public health. We evaluated our social media and assessed audience segmentation to promote health equity for key populations (adults 65 and over, children under 5, immunocompromised individuals, pregnant people, and racial and ethnic minorities).

Evaluation Methods and Results: We first assessed overall social media metrics for food safety posts on Twitter, Instagram, and Facebook channels from January to September 2022. Certain food safety messages had high negative engagement, so we conducted A/B message testing on social media for two messages addressing the negative comments. One message discussed the scientific reason to avoid a risky behavior, and the other outlined how to do that behavior more safely. Analysis showed that the former post resulted in higher reach and engagement than the latter, however, this engagement was more negative and more likely to contain misinformation.

We asked questions on a nationally representative survey to assess social media usage among key populations and preferred sources for receiving food safety messaging (n = 6594). Data showed that Americans from racial and ethnic minority groups were more likely to use Instagram, YouTube, and TikTok social media platforms than White Americans. For example, the odds of Instagram use were higher among Black Americans than White Americans (aOR = 1.45, 95% CI [1.16, 1.81]). Social media usage was far less likely among Americans aged 65 or older compared to those younger than 65. We also conducted 18 focus groups to gather feedback on food safety content from key populations. Participants responded positively to content that encouraged them to protect their loved ones and preferred messages emphasizing empowerment, self-efficacy, and individual choice to practice food safety. When we tested social media messages, participants preferred messages with graphics but responded negatively to those perceived to misalign with the written message (in this case, a child appearing happy accompanying a serious food safety message). Among participants living in a food desert and with low socioeconomic status (FD/LSES), Black and pregnant individuals reported food-related social media content consumption more frequently than non-FD/LSES counterparts. Non-FD/LSES Hispanic populations also reported consuming higher rates.

Conclusions: Social media is a promising tool for disseminating food safety information. Data revealed that Instagram may be effective to reach many key populations. Alternative communication channels may be better for adults aged 65 or older. Despite lower engagement rates, posting harm reduction food safety messages may be more useful for relevant topics. Special considerations must be made to ensure that messages remain effective when aiming to reach certain populations.

Implications for research and/or practice: Audience segmentation is useful for public health professionals to create targeted food safety content on social media. Further research should explore the effectiveness of social media messages among key populations using audience segmentation.